J O U R N A L O F M A T E R I A L S S C I E N C E 3 7 (2 0 0 2 ) 989 – 995
Quantification of the recrystallization behaviour in Al-alloy AA1050 S. P. CHEN, D. N. HANLON, S. VAN DER ZWAAG Lab. for Materials Science, Netherlands Institute for Metals Research, TU Delft Rotterdamseweg 137, 2628AL, Delft, The Netherlands E-mail:
[email protected] Y. T. PEI, J. TH. M. DEHOSSON Department of Applied Physics, Materials Science Center and Netherlands Institute for Metals Research, University of Groningen, Nijenborgh 4, 9747 AG Groningen, The Netherlands A new methodology for the determination of the recrystallized volume fraction from anodically etched aluminium alloys using optical microscopy is described. The method involves the creation of a composite image from multiple micrographs taken at a series of orientations. The results of quantitative analysis of images obtained by this new method are compared with those obtained using the traditional single image optical microscopy technique, orientation imaging microscopy (OIM) and microhardness indentation. The multiple orientation image method is shown to consistently yield a recrystallized volume fraction which is significantly higher than that determined from a single image while multiple orientation imaging and OIM results are found to be in good agreement. Furthermore it is shown that, after the subtraction of the effect of concurrent recovery using the rule of mixtures, microhardness indentation can also be used to determine the C 2002 Kluwer Academic Publishers recrystallized volume fraction.
1. Introduction In any study of the kinetics of recrystallization, it is important to determine the volume fraction of recrystallized material as accurately as possible. Several techniques are currently in use to quantify the recrystallization behaviour of deformed metals. The most widely used method is that of optical microscopy performed on a series of samples recrystallized to different extents [1–4]. In addition to yielding quantitative information regarding the extent of recrystallization, this technique provides some insight into other microstructural features such as grain size as well as patterns of nucleation and growth. However, when this method is applied to aluminium alloys such as AA1050, it turns out to be rather difficult to obtain precise data on the recrystallized fraction. Traditionally, an anodic etching technique is used to reveal the grain structure and a number of micrographs are taken of randomly selected areas. A point-counting technique is then applied to obtain average values of the recrystallized fraction [1, 3]. However, when observed directly at a single orientation with respect to the polarisors the grain structure of a partially recrystallized aluminium sample prepared in this manner is rarely clearly and unambiguously visible in its entirety [5]. When the sample is rotated under polarized light, the microstructure in a field of view seemingly undergoes a metamorphosis in which apparently unrecrystallized regions begin to appear recrystallized (as previously hidden boundaries become visible) and apparently recrystallized regions begin to appear unC 2002 Kluwer Academic Publishers 0022–2461
recrystallized (as internal substructural details become visible). This situation leads to extra complications in the identification of recrystallized grains with only a single micrograph. In order to eliminate such uncertainties in the determination of the fraction recrystallized, in this paper, a new method has been designed which is based on the construction of a composite image of a single region produced from a set of single micrographs taken at a series of stage rotations. This method enables the structure to be faithfully revealed and thus enables the recrystallized fraction to be more accurately determined. Recently, orientation imaging microscopy (OIM) has been employed to determine the recrystallized fraction [6–9] in Al-base alloys. Although so far OIM has been predominately applied to the determination of texture, grain boundary structure and phase determination, an important potential application of OIM lies in the field of recrystallization, in particular the determination of recrystallization kinetics and the crystallographic relationships between recrystallized and unrecrystallized grains. By OIM, it is possible to determine accurately whether an area is recrystallized. Other, indirect methods (hardness indentations, x-ray diffraction, neutron diffraction and electrical resistivity) have also been employed [1, 10–12] to determine the recrystallization fraction. These methods measure certain effects of micostructural changes on the properties and provide only average values including both recovery and recrystallization effects. Nevertheless, if 989
the effects of recovery and recrystallization can be separated such techniques may provide valuable additional information regarding recrystallization behaviour. In this study OIM and microhardness measurement have been employed along with the optical microscopy techniques described above to determine the recrystallization fraction in the commercially pure AA1050 alloy. The results so obtained are compared and critically discussed. 2. Experimental details 2.1. Material preparation The chemical composition of the commercially pure aluminium alloy AA 1050 used is: 0.185 wt% Fe, 0.109 wt% Si and Al in balance. The material contains plateshaped intermetallic compounds FeAl3 , which have an aspect ratio in the range of 1 to 6. The size of FeAl3 particles ranges between 0.2 and 7 µm. In order to produce material in a suitable form for further experimentation a cast ingot of AA1050 was hot rolled in 19 passes, resulting in a reduction in thickness from 500 mm to 4 mm. The hot rolling processes started at 520◦ C and finished at 305◦ C. The hot-rolled material was annealed at 600◦ C for 2 h and then quenched into water. The material was again heated to 400◦ C and held at this temperature for 2 h to reduce the content of iron in solid solution. The average grain size after this treatment was 90 µm. The materials were finally cold rolled to a reduction in thickness of 50% (from 4 mm to 2 mm). The rolled sheet was cut into small samples of 20 × 15 × 2 mm. The samples were annealed in a salt bath at 340◦ C for times ranging from 30 s to 3 h and then quenched into water to obtain different extents of recrystallization. The temperature of the salt bath was controlled to within an accuracy of ±2◦ C. The time taken for a specimen to reach the set temperature was approximately 5 seconds.
2.2. Microstructural characterization The optical microscopy and OIM as well as microhardness examinations were carried out on the section parallel to the RN plane (R rolling direction and N surface normal) to encounter as many grain boundaries as possible. After standard sectioning and polishing, specimens for optical metallography were etched anodically with Baker’s reagent (1% HBF4 aqueous solution) [13] at 20 V for approximately 120 s depending on the annealing time. For OIM scanning, the specimens were etched with Keller’s reagent for 30 seconds. In order to account for inhomogeneities in the microstructure in the through thickness direction of the sheet, all the metallography and hardness measurements were performed at locations along the center line of the sheet, in a band covering approximately one third of the thickness of the sheet. The optical microscopy examinations were performed using a NEOPHOT inverted stage metallurgical microscope. Both a standard point-counting technique and a LEICA QUANTIMET digital image analysis facility were used to evaluate the recrystallized frac990
tion. In all cases measurements were conducted on five separate areas. Orientation imaging microscopy (OIM), developed by TexSEM Laboratories Inc., was integrated with a Philips XL-30s FEG scanning electron microscope (SEM) and employed for the combined microstructural and crystallographic analysis. By a fast procedure of capturing and processing electron back-scatter diffraction patterns (EBSPs), the OIM system produces thousands of orientation measurements, linking local lattice orientation with grain morphology. Each measurement is represented by a pixel in the orientation micrographs, to which a colour or grey scale value is assigned on the basis of the local details of lattice orientation or the quality of the corresponding EBSP image quality (IQ). Recrystallized grains are distinguished according to IQ and orientation spread. The volume fraction of recrystallized material was determined from the ratio of the area of the recrystallized grains to the area of the whole image. In this work the area of a typical OIM scan was about 1800 × 600 µm2 , containing approximately 130 deformed grains. A Buehler OMNIMET MHT automatic micro hardness tester was used for the microhardness measurements. The micro hardness of each specimen was determined, using 50 g load and 15 s loading time. Hardness tests were made after polishing and 18 measurements were taken on each specimen.
2.3. Composite image method In order to construct composite images a total of four micrographs at 20◦ stage rotation intervals were taken from each area. The rotation centre of the stage should be aligned perfectly with the objective lens so that no misalignment of the micrographs results. The following step was to trace all visible recrystallized grains from each of the micrographs onto an acetate sheet in order to construct a single representation containing all recrystallized regions. To reveal the contribution to the total volume fraction made by each rotation, a separate colour was used to trace the visible recrystallized structure from each individual micrograph. The border of each micrograph was also traced in order to define a common overlaid area. Quantitative metallography (a standard point counting technique) was then performed using the composite image. In this study the criteria used to determine whether a grain is recrystallized were the following: 1. Size: Grains with an area considerably smaller (
